kootenpv / whereami

Uses WiFi signals :signal_strength: and machine learning to predict where you are
GNU Affero General Public License v3.0
5.1k stars 247 forks source link

add a time dimension to locations? #44

Open resistor4u opened 6 years ago

resistor4u commented 6 years ago

Ambient conditions (fog, storms, air quality, etc.) regularly affect radio transmissions in the suburban neighborhoods I've lived at. Some ambient conditions--like storms and fog--occur less regularly than the nighttime drops in temperature and humidity. (On the average, for any given geographic location, temperature and humidity are correlated by time of day.) The influences of ambient conditions on wifi RSSI will be less pronounced in denser, urban environments than suburban environments. The same goes for more climate-controlled environments (large libraries, universities, etc.). So, this may not be applicable in all use cases.

Nonetheless, given this time constant, I think factoring time-of-day into measurements and samples would be an interesting and useful addition to the project. I've kinda implemented this by changing the names of my classes, e.g., "shop-bedroom-day" and "shop-bedroom-night" or "shop-office-day" and "shop-office-night."

Is there a way the project could make this more formal? Like, adding a time-of-day component to locations?